In principle, highly ?uorescent rhodopsins could be employed as voltage sensors in neuroscience imaging. However, rhodopsin mutants featuring a ?uorescence ef?ciency close to the one of the green ?uorescence protein have not been discovered yet. Most importantly, we lack the molecular-level understanding required to predict if and how a speci?c set of mutations can amplify the weak ?uorescence of the rhodopsin chromophore. We propose to exploit state-of-the-art quantum chemical methods to systematically investigate these issues and to reengineer a microbial rhodopsin into a highly sensitive, genetically encodable, optogenetic tool for action potential visualization. -1-